One new and specialized type of imaging involves the capture of low intensity light—often on the order of only tens to hundreds of photons—from a light emitting sample. The source of the light indicates portions of the sample, such as traced molecules in a particular portion of a laboratory mammal, where an activity of interest may be taking place. For example, specialized in-vivo imaging applications may include analysis of one or more representations of emissions from internal portions of a specimen superimposed on a photographic representation of the specimen.
Digital cameras used in these imaging systems output image data in “analog digitizer units” (ADU) or “counts”. Counts are an arbitrary unit produced by the camera based on the amount of light it receives. The camera includes an array of pixels, each of which converts photons to electrons and generates digital output relative to the number of photons incident on the pixel. A digitizer detects the number of counts produced by the pixels and may add a gain to produce a more suitable output signal. Data from the camera thus comprises a digitized number of counts that is proportional to the number of photons incident on the pixels.
Counts are not an absolute indication of light activity for the light source. Imaging conditions and hardware used in the imaging system for a particular image affect count output. For example, hardware features such as a filter, an iris, f-stop, or position of the light source all affect the number of photons received by the camera, and thus the number of counts output by the camera. In addition, different camera designs may also affect data output. For example, camera output may be altered by a gain that turns pixel output into a more suitable number, which may vary from camera to camera. Counts are thus relative units that refer to the amplitude of a signal output by the camera.
A problem with using counts as a basis for imaging data analysis is that imaging data may only be analyzed or compared locally within a system, and often under identical imaging conditions. It is often desirable to compare imaging data taken under different imaging conditions (e.g. with a different lens) or with different imaging systems (e.g., imaging systems in different locations or using different hardware). Such imaging systems are not currently available. In View of the foregoing, techniques for obtaining absolute imaging data in a low-level light imaging system would be desirable.